JOINT COMMENTS OF THE AMERICAN BAR ASSOCIATION SECTION OF ANTITRUST LAW AND SECTION OF INTERNATIONAL LAW

Transcription

1 JOINT COMMENTS OF THE AMERICAN BAR ASSOCIATION SECTION OF ANTITRUST LAW AND SECTION OF INTERNATIONAL LAW RESPONSE TO REQUEST BY THE EUROPEAN COMMISSION FOR PUBLIC CONSULTATION ON DRAFT REGULATIONS AND GUIDELINES FOR THE ASSESSMENT OF CO-OPERATION AGREEMENTS BETWEEN COMPETITORS June 2010 The views expressed herein are presented on behalf of the Section of Antitrust Law and the Section of International Law only. These Comments have not been approved by the House of Delegates or the Board of Governors of the American Bar Association and should not be construed as representing the policy of the American Bar Association. INTRODUCTION The Section of Antitrust Law and the Section of International Law of the American Bar Association (the Sections ) respectfully submit these comments in response to the request issued by the European Commission (the Commission ) for public consultation on the following draft regulations and guidelines for the assessment of horizontal co-operation agreements, i.e.: Draft Commission Regulation (EU) No. /.. of [ ] on the application of Article 101(3) of the TFEU to categories of specialisation agreements ( Draft Specialization BER ); Draft Commission Regulation (EU) No. /.. of [ ] on the application of Article 101(3) of the TFEU to categories of research and development agreements ( Draft R&D BER ); and Draft Guidelines on the applicability of Article 101 of the TFEU to horizontal cooperation agreements ( Draft Horizontal Guidelines or Draft Guidelines ).

2 The Sections appreciate this opportunity to comment on the draft regulations and guidelines. These comments expand upon the Sections comments submitted in January 2009, in response to the questionnaire issued by the EC. 1 The Sections again wish to make clear that they appreciate and generally support the goal of the draft regulations and guidelines to provide an analytical framework for horizontal agreements grounded in economic analysis. This approach generally promises to provide business needed flexibility to fashion cooperative agreements within the framework of the EC competition rules. The draft regulations and guidelines also offer the potential for greater harmonization or convergence between the U.S. and EU competition laws. The convergence of competition law on the basis of sound economic principles remains an important objective to enable effective oversight of competitive markets at all times, including times of economic challenge. The draft regulations and guidelines are also significant because they recognize the practical requirements for business to cooperate in a market place increasingly characterized by globalization, rapid advancements in technology, and fluctuating market dynamics. The Sections hope that these comments will provide a helpful perspective based on experience both with U.S. antitrust laws and competition laws in the EU, its Member States and other competition regimes worldwide. The Sections would be pleased to provide any additional comments, or to participate in further consultation, as appropriate. 1 See January 2009 Comments of the American Bar Association s Section of Antitrust Law and Section of International Law in Response to Questionnaire Issued by the European Commission in connection with its review of the current regime for this assessment of horizontal co-operation agreements (hereinafter Joint Comments ), available at

3 EXECUTIVE SUMMARY I. Research and Development Agreements Articles 3(2) and 3(4) of the Draft R&D BER mandate certain approaches to intellectual property disclosure and access to know-how which may not be appropriate for all R&D agreements and may create disincentives for parties to enter into R&D agreements. The Sections suggest that private parties be given greater flexibility to address these issues based upon the facts and circumstances of particular collaborations. Moreover, as noted in the January 2009 comments, Article 4(1) of the current R&D BER effectively requires parties to dissolve successful R&D ventures after seven years, thereby creating potential disincentives to forming these types of ventures; this concern remains in the new Draft R&D BER. The Sections believe that the fruits of R&D ventures should be subject to a block exemption as the product of a continuing cooperation between non-competitors. 2 Finally, the Sections reiterate their suggestion that the market share thresholds in Articles 4(2) and 4(3) are too low. II. Agreements on Standards The primary comments offered by the Sections in relation to the Commission s approach to standard setting include: (i) the suggestion of additional introductory language expressing the need to balance concerns over potential anticompetitive effects from such activities with the goal of preserving appropriate incentives for investment in research and development; (ii) a recommendation that the Commission adopt less prescriptive language when guiding standard setting organizations (or SSOs) in how to deal with the issue of patent hold-up; and (iii) related 2 See p. below

4 suggestions concerning the rules governing ex ante joint negotiations over royalty rates or other licensing terms. 3 III. Information Sharing The Sections commend the Commission for including in the Draft Guidelines an extensive and thoughtful explanation on how information sharing agreements will be evaluated. The Sections encourage the Commission, however, to reconsider its proposed before-and-after analysis of the competitive effects of information exchanges, particularly in light of the practical difficulties that would arise in trying to measure changes in the level of market transparency. The Sections also suggest that the Commission consider providing additional explanation regarding the extent to which parties exchanges of information might be used as evidence of an agreement subject to Article 101. IV. Joint Purchasing Agreements The Sections comments respecting joint purchasing agreements recommend that the Commission consider clarifying how the Draft Guidelines will be applied to joint purchasing arrangements that involve restraints that arguably could be characterized as hardcore restrictions under the Vertical BER; as well as whether there are some circumstances in which the operations of parties to a joint purchasing arrangement are so integrated as to obviate the application of the Vertical BER and the Vertical Guidelines. The Sections also suggest that the Commission should explain the extent to which (if at all) it may analyze the competitive effects of joint purchasing agreements in markets other than the relevant purchasing and selling markets, even if only in certain sectors (e.g., insurance). 3 See p. below

5 V. Draft Specialization BER and Joint Production Agreements The Sections have suggested three technical changes to the Draft Specialization BER that would clarify its application to unilateral specialization agreements; open the door for parties to reciprocal specialization agreements to achieve added efficiencies; and allow for the continued application of the BER to agreements between or among parties whose combined market shares only temporarily exceed the safe harbor thresholds, but quickly recede below the level at which their agreements plausibly could cause any harm to competition. COMMENTS I. Research and Development Agreements The Sections commend the Commission for the Draft R&D BER s recognition that research and development ( R&D ) agreements make positive contributions in the great majority of circumstances and in only relatively isolated circumstances could give rise to competitive concerns. The Sections have some limited comments regarding Articles 3(2), 3(4), 4(1), 4(2) and 4(3) of the Draft R&D BER. A. Disclosure Article 3(2) of the Draft R&D BER provides that parties must agree that prior to starting the research and development all parties will disclose all their existing and pending intellectual property rights in as far as they are relevant for the exploitation of the results by the other party. This is a new requirement. The Sections are concerned that this provision may cause unnecessary disputes and create disincentives for parties to enter into R&D agreements. It will frequently be difficult or impossible for parties to know ex ante which intellectual property rights may be relevant to the exploitation of the results of a research project. Moreover, the concept of relevance is inherently vague

6 The Sections respectfully suggest deleting or significantly revising this provision. The provision appears directed at avoiding a situation where one party to an R&D agreement is unable to exploit the results of the agreement because the other party to the agreement asserts blocking intellectual property rights. The parties to R&D agreements have strong incentives to address this issue themselves in the governing agreement. There are a variety of ways that parties may seek to address this potential risk. Some parties might elect to address this risk through ex ante disclosure obligations of the type included in Article 3(2). Other parties might elect to address the issue through licensing provisions or other means, however. The Sections believe that parties to commercial arrangements of this kind have ample capacity and incentives to protect themselves from concerns of the type addressed by Article 3(2), and that it would be unduly restrictive to mandate one particular approach to this issue. Whether it is necessary or advisable to invoke antitrust principles as distinct from the law of deception or other similar legal regimes -- to address problems that may arise where this process goes awry is still a matter of intense debate among scholars and practitioners. While the Sections take no position on this debate at this time, the Sections recommend that, before committing to the specific approach proposed in the draft Guidelines, the Commission should consider and articulate its views on this issue. B. Access to and Right of Exploitation In the January 2009 comments, the Sections noted a concern that the current Article 3(3) of the R&D BER is ambiguous in its scope, because it can be interpreted to provide that each party must be free independently to exploit any pre-existing know-how of the parties necessary for the exploitation of the results of the joint research and development, even if that party neither contributed to, nor used, the pre-existing know-how in the R&D effort. The Sections suggested - 6 -

7 that any revisions should clarify that access to preexisting know-how is limited to know-how contributed to the joint project unless otherwise agreed by the parties. Article 3(4) of the Draft R&D BER appears to resolve the ambiguity in the current R&D BER, but in a fashion that may create unnecessary disputes and discourage parties from entering into R&D agreements. The new Article 3(4) provides that R&D agreements must stipulate that each party must be granted access to any pre-existing know-how of the other parties, if this know-how is indispensable for the purposes of its exploitation of the results. This provision appears directed at avoiding a situation where one party to an R&D agreement is unable to exploit the results of the agreement because the other party has proprietary know-how of some sort. The parties to R&D agreements have strong incentives to address this issue themselves in the governing agreement. There are a variety of ways that parties may seek to address this potential risk. Some parties might elect to address this risk through an ex ante commitment of the type included in Article 3(4). Other parties might elect to address the issue though other means, however. The Sections believe that it is unduly restrictive to mandate one particular approach to this issue. Moreover, a broad obligation to share any know-how which is deemed ex post to be indispensable may discourage parties from entering into R&D agreements. As noted in the January 2009 comments, the Sections suggested that access to preexisting knowhow should be limited to know-how contributed to the joint project or otherwise agreed by the parties. C. Duration In the January 2009 comments, the Sections noted that Article 4(1) of the current R&D BER unnecessarily introduces uncertainty into the continued viability of a R&D venture that has delivered substantial benefits. The new Draft R&D BER does not address this issue. As noted in the January 2009 comments, Article 4(1) of the R&D BER provides, in effect, that two parties - 7 -

8 that do not compete but who jointly develop and then exploit the product of their joint R&D lose the benefit of the block exemption for the joint R&D if they are successful in marketing their products for seven years. Article 4(1) of the R&D BER thus effectively requires non-competing parties to a joint R&D project to plan for the eventual dissolution of the venture if it is successful. In this regard, Article 4(1) of the R&D BER detracts from the long-term value of R&D agreements and imposes potential arbitrary dissolution costs, thereby creating disincentives that may discourage their formation in the first instance. The Sections suggested the R&D BER be revised to treat the R&D as the product of a continuing cooperation between noncompetitors as described in 124 of the Draft Horizontal Guidelines and thus the block exemption would continue to apply without limitation. The Sections respectfully reiterate this suggestion. D. Safe Harbor Threshold In the January 2009 comments, the Sections submitted that the 25% threshold of Article 4(2) and (3) of the R&D BER is too low, given that a share of this magnitude generally will not give rise to substantive competition law concerns. 4 The new Draft R&D BER maintains these thresholds. The Sections hereby incorporate by reference its earlier comments and respectfully suggests that a higher threshold (on the order of 35%) would be more appropriate. II. Agreements on Standards The clear benefits of standard-setting have led the Commission and European Courts to adopt a generally favorable view of standards under Community law. Consistent with this, the treatment of standardization agreements in the Draft Horizontal Guidelines ( ) sets out a positive framework for standard-setting in the context of formal standard-setting bodies. The 4 The Sections understand that the R&D BER and the Specialization BER, discussed below, would not apply to naked restraints of trade, but only to collaborations that involve some degree of integration

9 Sections suggest that the Guidelines could, however, express a more positive acknowledgement of the value of research and development, recognize the desirability of preserving the flexibility and market-based results of private arms-length negotiation, and endorse competition enforcement in particular and well-defined circumstances. At the same time, the Sections note the Guidelines detailed prescription of the scope of FRAND licensing standards and formulas for assessing their application. Concern for the interests of standards users and the desire to control possible anticompetitive conduct by IP owners (e.g., through deceptive practices and withholding of information that may create the potential for hold-up ) are legitimate parts of the competitive assessment of the structure and activities of standard-setting bodies. There is an ever-present need, however, to remain sensitive to the possibility that, despite best intentions to the contrary, intervention by competition agencies in private negotiations could undermine the development of workable and effective market-based solutions to the determination of licensing terms. To preserve essential long-run incentives for innovation and the dedication of valuable IP for the use of standard-setting bodies, competitive assessment should maintain a high regard for the value of intellectual property and the contribution of inventors. As explained in greater detail below, the Commission may wish to highlight this balancing aspect in order to avoid suggesting that prescription of licensing terms and other IP-related activities of standard-setting bodies is undertaken routinely or without concern for potential adverse long-run effects on innovation and IP. The Sections offer the following suggestions for the Commission s consideration, in part, to further the on-going efforts at convergence in antitrust enforcement. A. Application of Article 101 to Standard-Setting Agreements Generally First, the Sections note that the Draft Guidelines, while acknowledging that standardization agreements generally have a positive economic effect ( 258), have expanded - 9 -

10 significantly from the existing Horizontal Guidelines the discussion of the potential sources of anticompetitive effects from standard-setting activities. 5 This additional discussion is regarded as a welcome addition to the Horizontal Guidelines, as it can provide valuable guidance as to the circumstances in which standard-setting can lead to competitive harm. There is also, as acknowledged elsewhere in the Draft Guidelines, the need to balance the potential for such anticompetitive harm with the pro-competitive benefits from the exploitation of intellectual property rights. 6 In the final analysis, the Draft Guidelines should preserve the flexibility of existing practice, and the fact of standardization should not create an exceptional situation that reduces the protections accorded intellectual property rights. To address this issue, the Sections make the following suggestions. First, the Sections suggest that the Commission consider including a brief prefatory statement in this chapter of the Draft Horizontal Guidelines emphasizing the benefits accruing from innovation through intellectual property and offering the policy perspective that a reasonable market-based return to IPR holders is critical to spurring such innovation. The Sections view is that a policy statement that lacks such explicit balance might be perceived as expressing hostility to high royalty rates for intellectual property, regardless of how the market position of the IPR holder might have been attained, which in turn can create disincentives to investment in research and development. 7 5 See See See, e.g., OECD, Background Note by the Secretariat, Roundtable on Competition, Patents and Innovation, DAF/COMP (2009)8, available at (at 15-48); see also Thomas O. Barnett, Ass t Att y Gen., Antitrust Division, U.S. Dep t of Justice, Interoperability Between Antitrust and Intellectual Property, Presentation to the George Mason School of Law Symposium, Managing Antitrust Issues in a Global Marketplace, Washington, DC (Sept. 13, 2006), available at

11 Second, the Sections suggest that the Commission make more explicit in the Draft Guidelines that the focus in applying Article 101 to standard-setting agreements will be on consumer welfare considerations. While Article 101 provides a basis for analyzing disputes between interested parties over competing claims to the profits of innovation, this chapter should address potential benefits from innovation, such as access to technology, lower downstream prices and the like, while also taking into account effects on price and output for consumers and the incentives for future innovation. Third, the Sections note that several of the paragraphs in the Draft Guidelines refer to the value of patents as a concept distinct from royalty rates that might be reached if a sale or license of the intellectual property were the subject of arms-length negotiations. For example, 284 states: the assessment of whether fees imposed for patents in the standard-setting context are unfair or unreasonable, will be based on whether the fees bear a reasonable relationship to the economic value of the patents. Certainly some of the methodology suggested by the Commission for assessing fair value may be market-based and therefore likely to maintain proper incentives to innovate. 8 However, the Sections recommend including some clarifying language that the methodology suggested by the Draft Guidelines does not signal an effort by the Commission to adopt a government-mandated level for FRAND rates or a disregard for the good faith efforts by participants to reach agreement on FRAND rates. Moreover, the Sections believe that it would be helpful to gain greater knowledge of the effects of the various possible methodologies for assessing FRAND rates on innovation policy and on incentives for investment in research and development by firms before suggesting formulas for assessing FRAND rates 8 See id. (comparing the licensing fees charged for the relevant patents in a competitive environment before lock in)

12 which could chill investment in research and development and fail to acknowledge the considerable risks inventors undertake in research and development efforts. The application of FRAND licensing terms should continue to evolve through market-based resolution in armslength bilateral negotiations with the availability of contract actions if negotiations are not successful. Fourth, the Sections note that the Draft Guidelines refer in their title only to Art. 101 TFEU, but in fact they do contain some limited guidance regarding application of Art. 102 in regard to standard setting. See (under the heading '7.3 Assessment under Article 101(1)'). The application of competition rules to alleged exploitative abuse of market power, in particular where IPRs and standard-setting are involved, is recognized as an issue of great complexity, difficulty and importance, and carries the risk of presuming that license rates and terms have been established in a non-competitive fashion. Accordingly, the Sections suggest that the Commission consider whether it might be preferable to deal with this subject in a context where the broader range of questions can be considered, related to each other and dealt with in a more integral fashion. B. Comments on Specific Paragraphs of the Draft Guidelines The Draft Guidelines set forth at the conditions in which standard-setting agreements fall outside the scope of Article 101(1). For these paragraphs, the Sections offer the following specific comments. Paragraph 278 of the Draft Guidelines states, in part, that: There should be no bias in favour or against royalty-free standards, depending on the relative benefits of the latter compared to other alternatives. The Sections interpret this statement as signaling that if a standard-setting organization adopted a rule requiring royalty-free licensing, that this would be necessarily anticompetitive. The U.S. enforcement agencies recognize competing considerations, suggesting

13 desirability of a royalty-free rule in some circumstances. 9 The agencies overall conclusion on this and related issues is as follows: Neither Agency advocates that SSOs adopt any specific disclosure or licensing policy, and the Agencies do not suggest that any particular disclosure or licensing policy is required. 10 The Sections recommend that the Commission add language clarifying that it is not prohibiting standard-setting organizations from adopting a policy of royalty-free licenses, nor encouraging such policies. To address the issue of patent hold-up, the Draft Guidelines propose several mandatory standard-setting practices. For example, 280 of the Draft Guidelines states that the rules of a standard setting organization must seek to avoid patent hold-up conduct through binding rules; 281 goes on to say that an SSO s IPR policy should require good faith [ex ante] disclosure of essential IPRs and reasonable efforts to identify such IPRs; and 282 states that the IPR policy should also require all such IPR holders to provide an irrevocable FRAND license commitment. While the Sections recognize the need to address the issue of patent holdup, the experience to date with application of competition principles to patent hold-ups leaves wide room for debate both as to the proper definition of a patent hold-up and as to whether or in what circumstances such a practice should be prohibited by competition rules, and with what remedial consequences. For example, it can be suggested that intervention with regard to patent hold-ups be limited to cases of an affirmative misrepresentation regarding the existence of essential IP reading on a developing standard. 11 It can also be suggested that where the hold-up issue is founded on misrepresentation, the relevant branches of the law of business torts may be 9 See U.S. Dep t of Justice & Fed. Trade Comm n, Antitrust Enforcement and Intellectual Property Rights: Promoting Innovation and Competition (2007) ( DOJ/FTC Report ), at Id. at See Dell Computer Corp., 121 F.T.C. 616 (1996); and In re: Union Oil Co. of Cal., Docket No.9305 (F.T.C. Mar. 4, 2003), available at

14 attracted, in preference to competition law. In view of the current divergence of thoughtful opinions, the Sections suggest that the Commission consider the adoption of less prescriptive language than the must or should language of Similarly, the Sections suggest that the Commission consider less prescriptive language in 286. That paragraph states that: there should also be a requirement on all IPR holders who provide a FRAND license commitment to take all necessary measures to ensure that any undertaking to which the IPR owner transfers its IPR is bound by that commitment. The Sections recognize that the failure to bind transferees to an owner s commitment can be the source of competitive and consumer harm. The FTC s N-Data complaint and consent order of last year present one such possible example. 12 Nevertheless, these Draft Guidelines may go further than necessary to address the issue. The Sections suggest addressing the issue in the Draft Guidelines without directing that there should be such a transfer requirement that is applicable to all SSOs. 287 provides comfort under EC law for standard setting organizations to require or allow IPR holders to disclose ex ante their most restrictive licensing terms, including the maximum royalty rates. The Sections interpret this statement in light of U.S. guidance, including not only the DOJ/FTC Report but also the DOJ s 2006 Business Review Letter for VITA and 2007 Business Review Letter for IEEE. 13 While a safe harbor for such disclosure could be desirable, there is a risk that SSOs would consider absence of such a provision itself an infringement of Article 101. The Draft Guidelines should dispel this concern by making it clear 12 See Negotiated Data Solutions LLC, FTC No (Decision and Order) (Jan. 23, 2008), 13 See Letter from Thomas O. Barnett, Asst. Att y Gen., U.S. Dep t of Justice, to Robert A. Skitol (Oct. 30, 2006), and Letter from Thomas O. Barnett, Assistant Att y Gen., U.S. Dep t of Justice, to Michael A. Lindsay (April 30, 2007),

15 that absence of a mandatory disclosure provision in SSO rules will not be deemed an infringement or evidence thereof. Moreover, even as a safe harbor, the Draft Guidelines go beyond the current practice of most SSOs which do address disclosure and encourage reasonable efforts to disclose known IP relating to the standard and encourage, but do not mandate, disclosure. In addition, the Draft Guidelines appear to exclude from the safe harbor any ex ante joint negotiation of licensing terms, including rates. 14 The Sections recommend the addition of a cautionary statement to 287 of the Draft Guidelines that ex ante disclosure of rates creates some potential risk of overt or tacit coordination among implementers, driving royalty rates below a level that would result from arms-length bilateral negotiation and below the level necessary to create appropriate incentives for invention. The Sections believe that greater guidance is needed to ensure that the proposals of participants in standard setting organizations not amount to a group boycott involving purchases of intellectual property rights, with the anticompetitive effect of lowering prices to monopsonistic levels. The related concern that the Sections raise is that such proposals have the potential to facilitate tacit or express monopsony coordination among users who might be disproportionately represented in the standard setting organization. 15 Nevertheless, the Sections support the Commission s apparent intention to consider ex ante joint royalty negotiations as a matter requiring the assessment of the particular 14 See 287 ( should a standard-setting organisation's IPR policy require, or allow, IPR holders to individually disclose their most restrictive licensing terms, including the maximum royalty rates they would charge, prior to the adoption of the standard this will not lead to a restriction of competition within the meaning of Article 101(1) as long as the rules do not allow for the joint negotiation or discussion of licensing terms in particular royalty rates. ) 15 See Hill B. Wellford, Counsel to the Ass t Att y Gen., Antitrust Division, U.S. Dep t of Justice, Antitrust Issues in Standard Setting, 2d Annual Seminar on IT Standardization and Intellectual Property, China Electronics Standardization Institute, Beijing, China (Mar. 29, 2007)

16 facts and circumstances involved, which is analogous to the approach taken in the U.S. on the basis of its own enforcement experience. 16 On the issue of a standard s inclusion of alternative technologies, the Sections believe that the Commission may wish to reconsider its transporting what is a generally prudent precaution in connection with patent pools into the general standards development context. 288 states that: inclusion in a standard of substitute technologies may limit inter-technology competition by excluding one potentially competing alternative technology from being included in another standard ; and so, [a]s a general rule, the inclusion of substitute technologies in a standard is likely to give rise to unlawful effects under EC law. In the context of a patent pool, the incorporation of two substitute technologies may thereby suppress intertechnology competition by requiring all licensees to accept a package license containing all patents covering both technologies. Even in that context, it is not obvious in the Sections view that such an inclusion of alternative technologies will inevitably be anticompetitive and will never be beneficial to licensees. 17 In any event, assuming the anticompetitive effects in the patent pool context, the same does not necessarily follow from the mere use or incorporation of substitute technologies into a standard. The Sections do not see, in that standard setting context, why such incorporation would necessarily prevent any such substitute technology from also being used in a competing standard. Accordingly, the Sections recommend that the Commission consider drawing the appropriate distinction in this respect between patent pools and standard setting. 16 See DOJ/FTC Report at 49-56, particularly at ( Given the strong potential for procompetitive benefits, the Agencies will evaluate joint ex ante negotiation of licensing terms pursuant to the rule of reason. ) 17 See DOJ/FTC Report at

17 The Draft Guidelines appear to suggest in 316 (example) that the failure of an IPR owner to declare a willingness to license on IP terms could itself violate Article 101. The Sections are of the view that, if their reading of this example is correct, the Draft Guidelines are creating a rule that is at odds with the general principle applied in both the U.S. and in the EU that an IPR owner has the right to choose the parties with whom and the terms and conditions upon which he is willing to transact business, except under the most exceptional circumstances. A FRAND undertaking in its current application constitutes an agreement to make a good-faith effort to reach an understanding on licensing terms and does not mandate compulsory licensing. The Sections also suggest that such a rule on mandatory licensing would amount to a departure from the policy with respect to licensing on FRAND terms set forth elsewhere in the Draft Guidelines. C. Standards Agreements Block Exemption The Sections note that the Commission has a mandate to issue a block exemption regulation concerning the application of standards. To increase legal certainty, the Sections would recommend that the Commission consider adopting such a block exemption regulation, based on the principles currently set forth in the Horizontal Guidelines (subject to the revisions developed through the current review process). Alternatively, the Commission may consider adopting one or more broader block exemptions concerning horizontal co-operation agreements into which it could incorporate such guidance for standards. III. Information Sharing The Sections recommended in their Joint Comments that further guidance from the Commission on the subject of information exchange would be beneficial. 18 Although evaluation 18 Joint Comments at

18 of information exchanges had already been addressed in the Guidelines on the Applicability of Article 81 of the EC Treaty to Maritime Transport Services ( Maritime Guidelines ), 19 the treatment was limited, and the Sections noted that there was ambiguity in how the exchange of publicly available information was evaluated under the Maritime Guidelines. 20 The Draft Guidelines thoroughly address the competitive effects of information exchanges ( ). The Draft Guidelines contain no safe harbors or market share thresholds for evaluating information exchanges. They outline the effects that various types of exchanges can have on competition, depending upon the structure of the market, the nature of the goods, means of communication, and other factors. Although expressed in general economic terms, the discussion is largely consistent with tests for evaluating the legality of information exchanges under U.S. antitrust law. 21 The Draft Guidelines focus on the effects of an information exchange undertaken pursuant to an agreement, a concerted practice, or a decision of an association of undertakings. ( 55, 60.) The Draft Guidelines also mention circumstances in which an information exchange may be evidence of the existence of an agreement, concerted practice, or decision of an association of undertakings a topic that repeatedly arises under U.S. antitrust law in the course of weighing whether evidence of an information exchange is a factor supporting an inference of conspiracy among firms in concentrated markets. 22 The Sections encourage and support the O.J. (C 245) Joint Comments at For an overview of the factors taken into account in evaluating information exchanges under the rule of reason, see generally ABA SECTION OF ANTITRUST LAW, ANTITRUST LAW DEVELOPMENTS (6th ed. 2007). 22 See, e.g., In re Coordinated Pretrial Proceedings in Petroleum Prods. Antitrust Litig., 906 F.2d 432, (9th Cir. 1990) (holding that evidence of exchange of pricing information supported inference of conspiracy under 1 of Sherman Act (15 U.S.C. 1)); In re Flash Memory Antitrust Litig., 643 F. Supp. 2d 1133, (N.D. Cal. 2009) (holding that allegations that Defendants routinely exchanged highly sensitive competitive information, including pricing and production data, were sufficient to support inference of conspiracy under 1); see also ABA (Continued)

19 Commission for providing this guidance on the type of conduct and restrictive effects that may be lead to information exchange being treated as a cartel (see 59, 62), but given the importance of the issue and the severity of the remedies applicable to cartel conduct the Sections encourage the Commission to provide even greater clarity regarding the conditions and circumstances that would lead the Commission to infer cartel conduct from an information exchange. The Sections note that 54 of the Guidelines describes how information exchange may occur through a third party (referred to as indirect exchange, see footnote 39). While the Sections accept that indirect exchange of information can result in a breach of competition law, the circumstances in which that might be the case are relatively narrow. Moreover, because information flows are so often beneficial in typical market interactions, including every-day procompetitive negotiations between buyers and sellers, the Commission should take care to avoid placing any cloud over such indirect information exchange. While a demonstrated intention to influence the ultimate recipient s behavior in an anticompetitive fashion may warrant enforcement concern (see e.g,. UK Court of Appeal, Hasbro (2006) and Replica Kits (2006)), the Commission should be cautious to specify the precise circumstances in which such indirect exchanges may be viewed as problematic. Certain types of information exchange fall outside the scope of the Draft Guidelines, including the exchange of individualised data on intended future prices or quantities, or other exchanges, which are considered as restrictions of competition by object under Article 101(1). (Continued) SECTION OF ANTITRUST LAW, PROOF OF CONSPIRACY UNDER FEDERAL ANTITRUST LAWS (2010) (collecting cases)

20 ( 69.) These types of exchanges, having as their purpose restriction of competition, may constitute cartels. ( 68.) The Draft Guidelines recognize that information exchanges, other than those that may be considered restrictions on competition by object, may often be procompetitive and must therefore be analyzed on a case-by-case basis. This approach entails a comparison of the likely effects of the information exchange with the competitive situation that would prevail in the absence of the information exchange ( 69) and takes into account that whether an exchange would have restrictive effects depends on both the economic conditions on the relevant markets and the characteristics of information exchanged (id.). The approach is essentially the same as rule-of-reason analysis used to evaluate the legality of information exchanges under Section 1 of the Sherman Act. 23 The Draft Guidelines provide that evaluation of the effects of an information exchange requires before-and-after analysis: [I]t is important to assess the restrictive effects of the information exchange in the context of both the initial market conditions, and how the information exchange changes those conditions. ( 70.) Noting that collusive outcomes are more likely in transparent markets, the Draft Guidelines stress at 74 that the lower the preexisting level of transparency in the market, the more value an information exchange may have in attaining a collusive outcome. The key element in evaluating the change in the level of transparency is to identify to what extent the available information can be used by companies to determine the actions of their competitors. ( 74.) This type of before-and-after analysis has 23 The Supreme Court noted in United States v. U.S. Gypsum Co., 438 U.S. 422 (1978), that the exchange of price and other information among competitors may be procompetitive and suggested that an evaluation of its competitive effects would entail consideration of the structure of the industry involved and the nature of the information exchanged. 438 U.S. at 441 n.16. For the most thorough judicial analysis of an information exchange under the rule of reason, see Todd v. Exxon Corp., 275 F.3d 191, (2d Cir. 2001) (evaluating exchange of salaries and employee benefits information among competitors) (Sotomayor, J.)

21 not been explicitly recognized in the case law under Section 1 in which information exchanges have been evaluated, and it is problematic as to how it could actually be conducted. The difficulties in applying before-and-after analysis in practice could be daunting, and success would depend upon near-laboratory conditions. It could be impossible to measure the change in transparency if, for example, an information exchange were of long duration and changing market conditions had made it impossible to isolate the effects of the exchange and separate them from effects resulting from other factors impacting the market. The Sections would recommend, therefore, that the Commission not tie evaluation of the competitive effects of an exchange so closely to measurement of change over time in the level of transparency in the market but, instead, identify this as one among multiple approaches to evaluating competitive effects. The Draft Guidelines recognize in 81 that exchange of current strategically useful data, such as pricing, customer lists, production costs, or marketing plans, is more likely to be caught by Article 101 than exchanges of other types of information, and there would appear to be no need to undertake before-and-after analysis if the facts showed that these types of highly sensitive data were being shared among direct competitors. Even if the exchange were not in furtherance of a pure cartel, the competitively sensitive nature of these types of information could be expected in virtually every case to facilitate a collusive outcome by increasing transparency in the market, reducing market complexity, buffering instability or compensating for asymmetry. ( 73.) In common with the approach followed elsewhere in the Draft Guidelines, an evaluation of collaborative action under Article 101(3) requires that efficiency gains resulting from an information exchange must be passed on to consumers to an extent that outweighs the restrictive effects on competition caused by the exchange. ( 96.) Unlike the criteria for

22 evaluating purchasing agreements under Article 101(3) ( 214), there is no requirement, however, that cost savings to consumers be demonstrated. IV. Joint Purchasing Agreements The Joint Comments submitted in January 2009 raised a number of issues concerning the treatment of joint purchasing agreements under the current Horizontal Guidelines. 24 The Draft Guidelines address some of those issues, and also change in some respects the analysis of purchasing agreements. The comments below address (A) the recommendations made in the Joint Comments and (B) a change in how Article 101(3) may apply to joint purchasing agreements. A. Changes Recommended in Joint Comments The Sections recommended in the Joint Comments that the Guidelines be revised to establish criteria for evaluating purchasing agreements regardless of the size of the participating firms. 25 Paragraph 116 of the Guidelines recognizes that such agreements between small and medium-sized enterprises ( SMEs ) are normally procompetitive, but the Sections noted that a purchasing agreement formed by SMEs may account for as great a share of a relevant market as an agreement formed by two large firms. The Draft Guidelines omit any mention of SMEs in the discussion of purchasing agreements ( ), and there is no text in the Draft Guidelines corresponding to 116 of the Guidelines. Consistent with the Sections recommendation, evaluation of purchasing agreements in the Draft Guidelines does not turn on the size of participating firms, and there is no carve-out of safe harbors or creation of other separate treatment for SMEs. The Sections 24 Guidelines on the Applicability of Art. 81 of the EC Treaty to Horizontal Cooperation Agreements, 2001 O.J. (C 3) Joint Comments at

23 commend the Commission for framing the Draft Guidelines evaluation of joint purchasing agreements around well-recognized economic criteria specifically, the market power of the parties to the agreement in the relevant purchasing and selling markets and their respective suppliers available alternatives. The Draft Guidelines also address the concern previously expressed by the Sections regarding how and when the Guidelines, the Block Exemption Regulation on Vertical Restraints ( Vertical BER ), and the Guidelines on Vertical Restraints ( Vertical Guidelines ) will be applied to purchasing arrangements that involve both horizontal and vertical agreements. 26 Paragraphs 190 and 191 explain that horizontal issues will be addressed first, followed by any vertical issues. One issue the Draft Guidelines do not address is how putative conflicts between the Guidelines and the Vertical BER and Vertical Guidelines may be resolved particularly those conflicts that may be based on policy choices that are not reflected in all of the documents. In the Joint Comments, for example, the Sections noted that the Vertical BER (specifically, Art. 4(e), as recently revised) excludes as a hardcore restriction supply agreements that restrict a supplier s ability to sell components as aftermarket spare parts to end-users or independent service providers where the buyer (an original equipment manufacturer) has not been authorized to service its equipment. 27 But that restriction arguably could be read to prohibit a joint purchasing venture from exclusively selling components only to the OEMs that established the venture. The Sections again respectfully suggest that the Commission consider providing additional guidance addressing this point. 26 Joint Comments at Id

24 Nor do the Draft Guidelines address the circumstances under which the operations of a joint purchasing venture and its members may be sufficiently integrated such that evaluation under the Vertical BER and Vertical Guidelines may be inappropriate. 28 Clarification in this regard could provide members of such ventures more freedom to structure their operations and thereby enhance their ventures procompetitive benefits. The Sections also note that the Draft Guidelines omit any discussion of how the competitive effects of horizontal buying cooperatives in markets other than those in which the cooperative s participants are active (such as those referenced in the Commission s business insurance sector inquiry) may be analyzed. 29 The Sections continue to believe that the Guidelines could be enhanced by explaining the extent to which such effects may be material (or not) to the analysis of joint purchasing or other horizontal agreements. B. Change in Evaluation under Article 101(3) Although the Sections did not consider the effect of a purchasing agreement on consumers, the Draft Guidelines introduce this as a criterion for evaluating whether a purchasing agreement satisfies Article 101(3). Under Article 101(3), Article 101(1) may be declared inapplicable to an agreement, concerted practice, or decision by associations of undertakings that satisfies specific criteria. For a purchasing agreement to qualify for Article 101(3), it must be shown that resulting efficiency gains have been passed on to consumers to an extent that outweighs the restrictive effects of competition caused by the agreement. ( 214.) This means that cost savings must be passed on to the participants customers in the form of lower prices in 28 Id. at Id. at

25 the selling markets. 30 The Sections believe that this is a sound approach to measuring the procompetitive effects of collaboration among horizontal competitors. V. Draft Specialization BER and Joint Production Agreements The Sections stated in their January 2009 response to the Commission s public consultation that it supported the goal of the Specialization BER and were unaware of any major issues impeding its practical application. 31 Both observations remain true today, and the Sections further note that the Commission s proposed revisions, if adopted, will improve the clarity of the Specialization BER. Nonetheless, the Sections response also raised five technical issues concerning the Specialization BER that warranted clarification. Two of those issues already have been resolved by the Commission s proposed revisions. 32 The Sections respectfully ask the Commission to consider addressing the remaining three issues, set out again below (albeit in abbreviated form), in order that it may provide the business and legal communities with the clearest possible guidance on the application of the Specialization BER. First, revised Article 2(1)(a) former Article 1(1)(a) of the Draft Specialization BER remains ambiguous as to whether the exemption is limited to unilateral specialization agreements between only two parties, particularly in light of the reference to two or more parties in revised Articles 2(1) and 2(1)(b). 33 The Sections recommend that the Commission clarify this point or explain why the exemption should not apply to unilateral specialization agreements among multiple parties. 30 Id. 31 Joint Comments at The revisions reflected in proposed Articles 1(15)(b) and 3(b) confirm that pure specialization agreements paired with joint distribution may be subject to the BER. (See Joint Comments at 9.) The deletion of the phrase of a production joint venture in proposed Article 4(b), first sub-bullet, removes the suggestion in former Article 5(2)(a) that parties to a joint production agreement must form a production joint venture (nowhere defined in the BER) before ancillary agreements respecting production levels may be subject to the exemption. (Id.) 33 Joint Comments at

26 Second, the Sections recommend that the Draft Specialization BER be revised to expressly exempt reciprocal specialization agreements under which one of the parties to the agreement agrees to supply the product concerned only where the capacity of the other party is fully utilized. 34 Such arrangements pose no more risk to competition than the long-term, exclusive arrangements that are permitted under the BER, but may enable parties to achieve added efficiencies through specialization. Third, the Sections recommend that proposed Articles 5(2) and 5(3) former Articles 6(2) and 6(3) of the Draft Specialization BER, which exclude from the exemption agreements between parties whose combined market share exceeds the specified 20 and 25 percent thresholds, be revised to allow the exemption to continue to apply if the parties combined market share falls below 20 percent within the two- and one-year periods specified in those provisions. The market share thresholds are a proxy for an agreement s potential to injure competition. There is no reason to withhold the exemption from agreements that pierce that threshold temporarily, but quickly fall back within the safety zone and no longer pose a threat to competition. CONCLUSION The Sections are grateful for the opportunity to provide their views on the draft regulations and guidelines for the assessment of horizontal co-operation agreements, and it hopes that the Commission finds these comments to be useful. 34 Joint Comments at